Run Grounding DINO locally with OpenEye

    Language-guided object detection for identifying objects by description. Powers natural language scene queries. OpenEye, the open-source perception CLI, downloads Grounding DINO, runs it on your CPU or GPU, and returns structured JSON you can pipe into scripts, agents, and robots.

    TaskObject detection
    CLI idgrounding-dino
    CreatorIDEA Research
    WeightsIDEA-Research/grounding-dino-tiny (~341 MB)
    HardwareCPU, NVIDIA CUDA
    Install extraopeneye-sh[grounding]

    1. Install

    pipx install "openeye-sh[grounding]"

    2. Pull Grounding DINO

    openeye pull grounding-dino

    3. Run inference

    Output is JSON on stdout, so it composes with jq and any other Unix tool.

    openeye run grounding-dino image.jpg --prompt "red cup"

    4. Serve it as an API

    Start a local server with REST and WebSocket endpoints for apps, agents, and live camera streams.

    openeye serve grounding-dino

    See the serve reference, REST endpoints, and Python SDK.

    Other models